📊 Full opportunity report: The Frameworks Can’t See the Thing That Matters: A Year of AI-Enabled Cyber Threats on ThorstenMeyerAI.com — validation score, market gap, and execution plan.
TL;DR
A year-long analysis shows AI is increasing cyberattack sophistication and democratizing advanced techniques. Traditional threat metrics no longer reliably distinguish high-risk actors, prompting a reassessment of security strategies.
A new analysis from Anthropic reveals that AI is significantly altering the landscape of cyber threats, making attackers more capable and rendering traditional threat assessment methods obsolete. The report, based on 832 banned malicious accounts, shows that AI is enabling less skilled actors to perform complex, high-risk activities, which shifts the threat landscape in ways that security teams are unprepared for.
Anthropic’s study examined 832 accounts banned for malicious activity between March 2025 and March 2026, mapping their techniques onto the MITRE ATT&CK framework. The findings show that AI is primarily used to automate attack preparation, such as malware creation, with 67.3% of actors employing AI for this purpose. More critically, AI is increasingly used for advanced post-breach activities like lateral movement and account discovery, with these activities rising sharply over the year. Notably, the use of AI for lateral movement grew from 33% to 56% of actors, indicating a shift toward deeper, more damaging attacks. The report highlights that AI’s role in facilitating complex tasks diminishes the traditional link between attacker skill and threat level. Both novice and experienced actors now perform similar techniques, making threat assessment based on technique diversity or tool usage unreliable. Instead, the most dangerous actors are distinguished by how they deploy AI during critical stages of an attack, particularly in the later, more operational phases. This trend suggests that the democratization of advanced attack capabilities is eroding the security community’s existing threat models.The frameworks can’t see the thing that matters
For decades, danger meant which techniques an attacker commands. A year of real AI-enabled attacks — 832 banned accounts mapped onto MITRE ATT&CK — shows that signal breaking, just as a new, harder-to-see one takes over.
A year of real misuse, mapped to the standard taxonomy
A window, not a census — these are the cases with enough detail to assess techniques thoroughly. Inside it, the risk level climbed fast.
WHAT WAS STUDIED
THE RISK CLIMB · MEDIUM-OR-HIGHER ACTORS
cybersecurity threat detection software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
“More techniques” stopped meaning “more dangerous”
The old heuristic: count the techniques, judge the tooling. AI dissolved it — because the model supplies the techniques either way. Watch the old signal fail, then watch what it misses.
Risk score vs. technique count
Two ways to read the same attacker. One is going blind. Press play.
AI-powered malware analysis tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Deeper into the attack — and into less-skilled hands
Across the year, AI use drifted from getting in toward acting once already inside — the operationally demanding stages that used to require an expert.
The attack lifecycle · where AI is now applied
The center of gravity moved right — toward post-compromise work.
network security monitoring devices
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
From “what they know” to “what they’ve built”
The report sorts the signals into three tiers — one dead, one fading, one durable.
Technique count & tooling
16 vs. 20 between novice and expert; platform doesn’t correlate. The model supplies the techniques either way.
Where in the lifecycle AI is applied
Concentrating on operationally demanding, post-compromise stages is a better signal — but it’s eroding as the whole population heads there.
The scaffolding around the model
Architectures that let the model chain stages and run with minimal human input. Not what they know — whether they’ve built a system that lets AI run the attack.
advanced intrusion detection system
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Fixing the map before the territory moves again
A taxonomy that can’t name the most dangerous behavior on the field will quietly mislead the people relying on it. The response runs in two directions.
Fed back into the models
The findings informed safeguards on the most capable models, built to detect & block some of what was observed:
- Blocking malware development
- Blocking mass data exfiltration
- Putting tools in defenders’ hands first (Project Glasswing)
Taking it to the source
Following the Verizon work, Anthropic says it’s in discussions with MITRE about how ATT&CK might evolve:
- A vocabulary for agentic orchestration
- Naming the scaffolding that makes a model an operator
- An interactive technique visualization on the Red blog
Reading it in proportion
- The 832 cases are a detailed subset, not the full population — the precise percentages are directional, not definitive.
- “More autonomous” is not “fully autonomous” — even the standout case needed human input at key moments, which is itself a place for defenders to intervene.
- This is one vendor’s window — the company with visibility into misuse of its own model, publishing what it found. The right thing to do with the data, and worth remembering as you read it.
Implications of AI-Driven Attack Capabilities for Cyber Defense
This shift fundamentally challenges established threat assessment practices, which relied on counting techniques and analyzing tooling to gauge attacker danger. As AI enables less skilled actors to perform sophisticated, high-impact activities, traditional indicators become less meaningful. Security teams must now reconsider how they identify and prioritize threats, focusing more on how attackers deploy AI during various attack phases rather than solely on technical complexity or toolsets. This change increases the risk of underestimating threats and underscores the urgent need for new detection and response strategies in an AI-augmented threat landscape.
Evolving Threat Landscape and the Role of AI in Cyber Attacks
For decades, cybersecurity threat assessment has centered on the number of techniques used and the sophistication of tools employed by attackers. This approach has helped distinguish high-risk actors from amateurs. However, recent developments indicate that AI has begun to disrupt this paradigm. The use of AI to automate attack preparation and execution has been observed in real-world incidents, with attackers increasingly relying on AI for tasks like malware creation, lateral movement, and account discovery. The trend reflects a broader shift toward AI-enabled attack automation, which lowers the barrier to executing complex operations and blurs the lines between novice and expert threat actors.
“Our analysis shows a significant increase in AI-assisted lateral movement and account discovery, indicating a move toward more damaging, post-compromise activities.”
— Anthropic Research Team
Unclear Impact of AI on Future Threat Dynamics
While the report provides strong evidence of AI’s current role in enhancing attack capabilities, it remains uncertain how these trends will evolve in the coming months. The extent to which attackers will adopt or develop new AI tools, and how defenders will adapt their strategies, is still unclear. Additionally, the full scope of how AI might enable even more sophisticated or autonomous attacks remains to be seen, as research into AI-driven threat evolution continues.
Next Steps for Cybersecurity in an AI-Driven Era
Security professionals will need to develop new threat detection models that focus on attack behaviors and AI deployment patterns, rather than solely on technical complexity. Organizations may also invest in AI-based defense tools designed to identify malicious AI activity. Further research is expected to explore how AI can be both a tool for attackers and defenders, shaping the future of cybersecurity strategies and policies.
Key Questions
How does AI make attackers more dangerous?
AI automates complex attack tasks such as malware creation, lateral movement, and account discovery, enabling less skilled actors to perform high-impact activities previously requiring expertise.
Why are traditional threat assessment methods no longer reliable?
Because AI allows attackers to perform sophisticated activities regardless of their skill level, making technique diversity and tooling poor indicators of threat level.
What should organizations do to defend against AI-enabled attacks?
Organizations should develop new detection methods focusing on attack behaviors and AI usage patterns, and consider deploying AI-based security tools.
Will AI make cyberattacks more autonomous?
While current evidence shows AI enhances attack automation, the development of fully autonomous attacks remains a future possibility that requires further monitoring.
How might threat assessment change in the future?
Future threat assessment will likely prioritize behavioral analysis and AI activity detection over traditional metrics like technique count or tool type.
Source: ThorstenMeyerAI.com
